Customers can place orders for services such as travel and accommodations from merchants in real-time over the web. These orders can be received and processed immediately. However, payments for the orders typically require more time to complete and, in particular, to secure the money being transferred. Therefore, merchants typically require the customer to provide assurances of payment in real-time while the order is being placed. As an example, a customer may input credit card information pursuant to a payment, and the merchant may verify the credit card information in real-time before authorizing the sale. The verification typically includes determining whether the provided information is valid (i.e., that a credit card number, expiration date, PIN code, and/or customer name match known information).
Upon receiving such assurances, a purchase order may be generated for the customer. The purchase order provides evidence of the order such as, for example, a purchase price, goods and/or services ordered, and the like. Later, an invoice for the order may be generated. While the purchase order is usually used to indicate which products are requested and an estimate or offering for the price, the invoice is usually used to indicate which products were actually provided and the final price for the products. Frequently, the purchase price as demonstrated by the invoice for the order is different from the purchase price as demonstrated by the purchase order. As an example, if a guest at a hotel initially orders a 3-night stay but ends up staying a fourth night, the total price of the purchase order may reflect a different total price than that of the subsequent invoice. Cases in which the total price of the invoice is different from the total price of the purchase order are difficult to track, especially in large enterprises accepting many orders daily (e.g., in a large hotel chain managing hundreds or thousands of hotels in a given country). The differences may cause errors in recordkeeping for enterprises.
As businesses increasingly rely on technology to manage data related to operations such as invoice and purchase order data, suitable systems for properly managing and validating data have become crucial to success. Particularly for large businesses, the amount of data utilized daily by businesses can be overwhelming. Accordingly, manual review and validation of such data is impractical, at best. However, disparities between recordkeeping documents can cause significant problems for businesses such as, for example, failure to properly report earnings to tax authorities.
Some solutions exist for automatically recognizing information in scanned documents (e.g., invoices and receipts) or other unstructured electronic documents (e.g., unstructured text files). Such solutions often face challenges in accurately identifying and recognizing characters and other features of electronic documents. Moreover, degradation in content of the input unstructured electronic documents typically result in higher error rates. As a result, existing image recognition techniques are not completely accurate under ideal circumstances (i.e., very clear images), and their accuracy often decreases dramatically when input images are less clear. Moreover, missing or otherwise incomplete data can result in errors during subsequent use of the data. Many existing solutions cannot identify missing data unless, e.g., a field in a structured dataset is left incomplete.
In addition, existing image recognition solutions may be unable to accurately identify some or all special characters (e.g., “!,” “@,” “#,” “$,” “©,” “%,” “&,” etc.). As an example, some existing image recognition solutions may inaccurately identify a dash included in a scanned receipt as the number “1.” As another example, some existing image recognition solutions cannot identify special characters such as the dollar sign, the yen symbol, etc.
Further, such solutions may face challenges in preparing recognized information for subsequent use. Specifically, many such solutions either produce output in an unstructured format, or can only produce structured output if the input electronic documents are specifically formatted for recognition by an image recognition system. The resulting unstructured output typically cannot be processed efficiently. In particular, such unstructured output may contain duplicates, and may include data that requires subsequent processing prior to use.
During the course of daily work, employees sometimes use their personal vehicles for travel. Many businesses offer refunds for travel-based expenses and, in particular, mileage-based expenses such as cost of fuel. Mileage allowance payments (MAPs) are payments that are refunded by enterprises for use of their personal vehicles. Because businesses offering MAPs pay for the travel expenses, they may be entitled to value-added tax (VAT) refunds that are paid for travel expenses abroad. Typically, such refunds are limited to refunds on taxes paid for particular travel expenses such as fuel expenses (as opposed to, for example, wear and tear to employee vehicles). To this end, employees of businesses offering MAPs submit expense reports indicating information related to travel expenses in order to provide the information needed to refund the employee and seek any applicable VAT refunds.
The refund requirements vary based on country regulations, car types, and the like. In some countries, the lower amount between the fuel expense and the mileage expense is refunded. For example, if the rate per mile results in a calculated expense of 180 pounds and the actual cost of fuel is 200 Pounds, mileage expense of 180 pounds should be submitted for VAT refund processing purposes. The variations and rules may result in difficulties keeping records, particularly when proof (i.e., receipts) of the transactions needed for submitting VAT reclaims are in the form of unstructured data.
It would therefore be advantageous to provide a solution that would overcome the deficiencies of the prior art.